All of lore.kernel.org
 help / color / mirror / Atom feed
From: Linhaifeng <haifeng.lin-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>
To: "Xie, Huawei" <huawei.xie-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Cc: "dev-VfR2kkLFssw@public.gmane.org" <dev-VfR2kkLFssw@public.gmane.org>
Subject: Re: [PATCH] cast used->idx to volatile
Date: Tue, 24 Mar 2015 15:43:37 +0800	[thread overview]
Message-ID: <551115A9.8060106@huawei.com> (raw)
In-Reply-To: <C37D651A908B024F974696C65296B57B0F3F3F2B-0J0gbvR4kThpB2pF5aRoyrfspsVTdybXVpNB7YpNyf8@public.gmane.org>



On 2015/3/24 9:53, Xie, Huawei wrote:
> On 3/24/2015 9:00 AM, Linhaifeng wrote:
>>
>> On 2015/3/23 20:54, Xie, Huawei wrote:
>>>
>>>> -----Original Message-----
>>>> From: Linhaifeng [mailto:haifeng.lin-hv44wF8Li93QT0dZR+AlfA@public.gmane.org]
>>>> Sent: Monday, March 23, 2015 8:24 PM
>>>> To: dev-VfR2kkLFssw@public.gmane.org
>>>> Cc: Ouyang, Changchun; Xie, Huawei
>>>> Subject: Re: [dpdk-dev] [PATCH] cast used->idx to volatile
>>>>
>>>>
>>>>
>>>> On 2015/3/21 16:07, linhaifeng wrote:
>>>>> From: Linhaifeng <haifeng.lin-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>
>>>>>
>>>>> Same as rte_vhost_enqueue_burst we should cast used->idx
>>>>> to volatile before notify guest.
>>>>>
>>>>> Signed-off-by: Linhaifeng <haifeng.lin-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>
>>>>> ---
>>>>>  lib/librte_vhost/vhost_rxtx.c | 2 +-
>>>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>>>>
>>>>> diff --git a/lib/librte_vhost/vhost_rxtx.c b/lib/librte_vhost/vhost_rxtx.c
>>>>> index 535c7a1..8d674d1 100644
>>>>> --- a/lib/librte_vhost/vhost_rxtx.c
>>>>> +++ b/lib/librte_vhost/vhost_rxtx.c
>>>>> @@ -722,7 +722,7 @@ rte_vhost_dequeue_burst(struct virtio_net *dev,
>>>> uint16_t queue_id,
>>>>>  	}
>>>>>
>>>>>  	rte_compiler_barrier();
>>>>> -	vq->used->idx += entry_success;
>>>>> +	*(volatile uint16_t *)&vq->used->idx += entry_success;
>>>
>>> Haifeng:
>>> We have compiler barrier before and an external function call behind, so we don't need volatile  here.
>>> Do you meet issue?
>>>
>> Tx_q is sometimes stopped when we use virtio_net. Because vhost thought there are no buffers in tx_q and virtio_net
>> though vhost haven't handle all packets so we have to restart VM to restore work.
>>
>> The status in VM is:
>> Mar 18 17:11:10 linux-b2ij kernel: [46337.246687] net eth7: virtnet_poll
>> Mar 18 17:11:10 linux-b2ij kernel: [46337.246690] net eth7: receive_buf
>> Mar 18 17:11:10 linux-b2ij kernel: [46337.246693] net eth7: vi->num=239
>> Mar 18 17:11:10 linux-b2ij kernel: [46337.246695] net eth7: svq:avail->idx=52939 used->idx=52939 num_free=18 num_added=0 svq->last_used_idx=52820
>> Mar 18 17:11:10 linux-b2ij kernel: [46337.246699] net eth7: rvq:avail->idx=36215 used->idx=35977 num_free=18 num_added=0 rvq->last_used_idx=35977
>> Mar 18 17:11:11 linux-b2ij kernel: [46337.901038] net eth7: dev_queue_xmit, qdisc->flags=4, qdisc->state deactiveed=0
>> Mar 18 17:11:12 linux-b2ij kernel: [46337.901042] net eth7: dev_queue_xmit, txq->state=1, stopped=1
>>
>> Why compiler barrier not take effect in our case? Is compiler barrier depended on -O3 option? We use -O2 option.
> compiler barrier always works regardless of the optimization option.
> I don't get your story, but the key thing is, do you check the asm code?
> If called from outside as an API, how is it possible it is optimized?
> there is only one update to used->idx in that function.


Do you mean rte_vhost_enqueue_burst also not need cast used->idx to volatile ? Why not remove it?

>>
>>>>>  	/* Kick guest if required. */
>>>>>  	if (!(vq->avail->flags & VRING_AVAIL_F_NO_INTERRUPT))
>>>>>  		eventfd_write((int)vq->callfd, 1);
>>>>>
>>>> --
>>>> Regards,
>>>> Haifeng
>>>
>>>
> 
> 
> .
> 

-- 
Regards,
Haifeng

  parent reply	other threads:[~2015-03-24  7:43 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-21  8:07 [PATCH] cast used->idx to volatile linhaifeng
     [not found] ` <1426925237-8312-1-git-send-email-haifeng.lin-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>
2015-03-23  2:29   ` Linhaifeng
2015-03-23 12:23   ` Linhaifeng
     [not found]     ` <551005DD.4060508-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>
2015-03-23 12:54       ` Xie, Huawei
     [not found]         ` <C37D651A908B024F974696C65296B57B0F3F14D3-0J0gbvR4kThpB2pF5aRoyrfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2015-03-24  0:59           ` Linhaifeng
2015-03-24  1:53             ` Xie, Huawei
     [not found]               ` <C37D651A908B024F974696C65296B57B0F3F3F2B-0J0gbvR4kThpB2pF5aRoyrfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2015-03-24  7:43                 ` Linhaifeng [this message]
2015-03-24 10:06                   ` Xie, Huawei
     [not found]                     ` <C37D651A908B024F974696C65296B57B0F3F4DDD-0J0gbvR4kThpB2pF5aRoyrfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2015-03-24 12:28                       ` Linhaifeng
2015-03-30  9:20                       ` Linhaifeng
2015-03-30 15:56                         ` Xie, Huawei
     [not found]                           ` <C37D651A908B024F974696C65296B57B0F417D44-0J0gbvR4kThpB2pF5aRoyrfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2015-04-01  7:50                             ` Thomas Monjalon
2015-04-01 14:40                               ` Xie, Huawei

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=551115A9.8060106@huawei.com \
    --to=haifeng.lin-hv44wf8li93qt0dzr+alfa@public.gmane.org \
    --cc=dev-VfR2kkLFssw@public.gmane.org \
    --cc=huawei.xie-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.